1. Field of the Invention
The present invention relates to Group II precursor compositions and their synthesis, and to a method of forming a Group II metal-containing films on a substrate by metalorganic chemical vapor deposition (MOCVD) utilizing such precursor compositions.
2. Description of the Related Art
Many materials are utilized in the form of thin films on substrates and are formed by vapor deposition techniques.
A number of Group II metal (Ba, Sr, Ca, Ma)-containing films fall in this cateaory. These encompass refractory thin film high temperature superconducting (HTSC) compositions including YBa.sub.2 Cu.sub.3 O.sub.x, wherein x is from about 6 to 7.3, BiSrCaCuO, and TlBaCaCuO. Other Group II metal-containing films include barium titanate, BaTiO.sub.3, and barium strontium titanate, Ba.sub.x Sr.sub.1-x TiO.sub.3 (BST), which have been identified as ferroelectric, photonic and electronic materials with unique and potentially very useful properties in thin film applications of such materials. Still other Group II metal-containing films include materials such as Ba.sub.x Sr.sub.1-x Nb.sub.2 O.sub.6, which is a photonic material whose index of refraction changes as a function of electric field and also as a function of the intensity of irradiated light. Additional Group II metal-containing films include Group II metal fluorides such as BaF.sub.2, CaF.sub.2, and SrF.sub.2, which are useful for scintillation detecting and coating of optical fibers, and Group II doped lanthanum manaanites, such as La.sub.1-x Ca.sub.x MnO.sub.3.
Many of the potential applications of these materials require their use in thin film (&lt;1000 .mu.m) coatings, or layer form. The films or layers may also be advantageously epitaxially related to the substrate upon which they are formed. Applications in which materials may need to be deposited in film or layer form include intearated circuits, switches, radiation detectors, thin film capacitors, holoaraphic storage media, and various other microelectronic devices.
Chemical vapor deposition (CVD) is a particularly attractive method for forming these layers because it is readily scaled up for production. Further, the electronic industry has extensive experience and an established CVD equipment base that can be applied to new CVD processes. In general, the control of key variables such as film stoichiometry and thickness, and the coating of a wide variety of substrate geometries is possible with CVD. Forming the thin films by CVD permits the integration of these materials into existing device production technologies. CVD also permits the formation of layers of materials that are epitaxially grown on substrates having close crystal structures and lattice parameters.
CVD requires that the element source reagents, i.e., the precursor compounds and complexes containing the elements or components of interest, must be sufficiently volatile to permit gas phase transport into the chemical vapor deposition reactor. The elemental precursor must decompose in the CVD reactor to deposit only the desired element at the desired growth temperatures. Premature gas phase reactions leading to particulate formation must not occur, nor should the source reagent decompose in the lines before reaching the reactor deposition chamber. When compounds are decomposed for deposition, obtaining optimal properties requires close control of stoichiometry that can only be achieved if the reagent can be delivered into the reactor in a controllable fashion. In this respect the reagents must not be so chemically stable that they are non-reactive in the deposition chamber.
Desirable CVD reagents, therefore, are fairly reactive and volatile. Unfortunately, for many of the materials described above, volatile reagents do not exist. Many potentially highly useful refractory materials have in common that one or more of their components are Group II elements, e.g., the metals barium, calcium, strontium, or maanesium, for which no or few volatile compounds well-suited for CVD are known. In many cases, the source reagents are solids whose sublimation temperature is very close to the decomposition temperature. Therefore, the reagent may begin to decompose in the lines during transport to the reactor, and it therefore is very difficult to control the stoichiometry of the deposited films from such decomposition--susceptible reagents.
When the film being deposited by CVD is a multicomponent substance, such as barium titanate or the oxide superconductors, rather than a pure element, controlling the film stoichiometry is critical to obtaining the desired film properties (optical and/or electrical properties). In the deposition of such materials, which may form films with a wide range of stoichiometries, the controlled delivery of known proportions of the source reagents into the CVD reactor chamber is essential.
In other cases, the CVD source reagents are liquids, or are dissolvable or suspendable in solvents to form liquid precursor compositions. Such liquid precursors are suitable for liquid delivery CVD. Liquid delivery CVD as a process has a number of desirable features in relation to other reagent delivery techniques, such as conventional bubbler delivery, liquids. Nonetheless, the delivery of liquid precursors into the CVD reactor (in the vapor phase) after their vaporization has proven difficult in many instances because of problems of premature decomposition and/or stoichiometry control.
Thus, while precursor liquid delivery systems present distinct advantages over conventional techniques, there is often some fraction of the precursor compound that decomposes into very low volatility compounds that remain in the vaporization zone. This deficiency is an important issue in the operation of CVD processes that use thermally unstable solid source precursors that undergo significant decomposition at conditions needed for sublimation. Such decomposition can occur in all reagent delivery systems that involve a vaporization step, including flash vaporizer liquid delivery systems, as well as more conventional reagent delivery systems that include bubblers and heated vessels operated without carrier gas.
Optimization of the conditions used in the vaporizer of precursor delivery systems can minimize the precursor decomposition in the vaporization zone, but virtually all solid precursors decompose to some extent near their vaporization temperature. Although the use of these precursors may be mandated by availability or economics, thermal decomposition should be minimized during gas-phase transport. Use of liquid precursors, however, can alleviate some of the decomposition and residue problems encountered with solid source delivery systems. The elimination of solid precursors can also reduce the formation of particles and improve the vaporizer mean-time to maintenance.
Despite the advantages of the liquid delivery approach (which include improved precision and accuracy for most liquid and solid CVD precursors and higher delivery rates), the foregoing deficiencies pose a serious impediment to widespread use of the liquid delivery technique for providing volatilized reagent to the CVD reactor in the full-scale manufacturing of electronic components.
The foregoing problems have specifically been experienced in the development of high-density memories using high dielectric constant and ferroelectric materials. In addition to high-density memories, ferroelectric materials are attractive candidates in a wide variety of solid state sensors and imaging devices, as a consequence of their pyroelectric and piezoelectric properties. Production worthy deposition modules are needed to realize the full potential of ferroelectric materials in evolving semiconductors. The preferred method for production of films of these ferroelectric materials is MOCVD, but at present a full compliment of stable liquid precursors is not commercially available for many ferroelectric thins films of interest, such as BaSrTiO.sub.3 and Sr.sub.2 BiTa.sub.2 O.sub.9 (SBT).
The vaporization of solid Group II precursors such as those used in the MOCVD of these materials typically undergo some decomposition during vaporization. Oligomerization or polymerization may accompany decomposition and reduce the effective transport rate of the precursors more rapidly than expected based upon the amount of decomposed material. This phenomenon can lead to inefficient transport, particle formation, decomposition residue, loss of film stoichiometry and process drift or irreproducibility. It is one of the reasons why many groups have designed their CVD tools to allow the bubblers to be loaded with a fresh charade of precursor prior to each run. See, for example, J. M. Zhana, J. Zhao, H. O. Marcy, L. M. Tonae, B. M. Wessels, T. J. Marks, and C. R. Kannewurf, Appl. Phys. Lett., 54, 1166 (1989).
Further, the Group II precursor materials do not have high vapor pressures, so all sections of the reactor between the vaporization point and any trap used to remove undecomposed precursor before the vacuum pump must be heated. This brings considerable added complexity and expense to the reactor design. In particular, cost and complexity rise steeply with increase of required temperatures from around 180.degree. C. to 240.degree. C. This is because elastomer vacuum seals cannot withstand temperatures above the 200-220.degree. C. range and therefore, metal seals must be used. The development of liquid precursors that can be vaporized and will not condense as solids on the reactor walls at lower temperatures would entail significant commercial advantages for the production of Group II element-containing films, such as BST.
Improved liquid delivery systems are disclosed in U.S. Pat. No. 5,204,314 issued Apr. 20, 1993 to Peter S. Kirlin et al. and U.S. Pat. No. 5,536,323 issued Jul. 16, 1996 to Peter S. Kirlin et al., which describe heated vaporization structures such as microporous disk elements. In use, liquid source reagent compositions are flowed onto the formations vaporization structure for flash vaporization. Vapor thereby is produced for transport to the deposition zone, e.g., a CVD reactor. The liquid delivery systems of these patents provide high efficiency generation of vapor from which films may be grown on substrates. Such liquid delivery systems are usefully employed for generation of multicomponent vapors from corresponding liquid reagent solutions containing one or more precursors as solutes, or alternatively from liquid reagent suspensions containing one or more precursors as soluble suspensions. Other methodoloaies, such as aerosol generation can be envisioned, as well.
The art continues to seek improvements in Group II metal source reagent compositions, and in liquid delivery systems for vapor-phase formation of advanced materials comprising Group II metal(s).
The invention described hereinafter involves the synthesis of novel liquid, or low melting solid, Group II metal precursors, since the existing Ba and Sr precursors are solids and generally associated with unacceptable levels of particle formation, and vaporizer or delivery tube clogging. As a result, such precursors thus do not fully satisfy the requirements of the CVD process for manufacturing semiconductor components.
Designing liquid or low melting point solid Group II CVD precursors represents a significant challenge. For example, the difficulties in obtaining suitable barium precursors stem from the highly electropositive character of barium with its large ionic radius. High coordination numbers are present in useful barium coordination complexes that have utility for CVD to produce barium-containing films. Such high coordination numbers have to be satisfied by mostly neutral ligands that coordinate weakly to the electropositive Ba center.
As a result, the Ba CVD precursors synthesized to date are predomingntly characterized by insufficient thermal stability. The lability of metal-ligand bonds in Ba metalorganics coupled with the large size of the metal center result in the tendency to aaaregate to form multinuclear Ba species. The formation of multinuclear species can occur during synthesis, affecting the volatility of the precursor, or during the CVD process, leading to particle formation, lowered precursor transport and film growth efficiency. Formation of multinuclear species also results in decreased solubility, which is an important parameter of the liquid delivery CVD process.
In prior art synthesis of barium precursors, it was difficult to form very stable and highly volatile complexes of barium. As a result, synthesis efforts have focused on achieving an appropriate mix of volatility and stability for a given precursor to facilitate its use in CVD applications. Similar issues and considerations are presented in the development of other Group II metal (e.g., Ca, Sr, and Ma) precursors.
In view of the foregoing, it is an object of the present invention to provide new Group II precursor compositions which are usefully employed in liquid delivery MOCVD processes to form Group II metal-containing films.
It is another object of the present invention to provide new Group II precursor compositions for forming ferroelectric films of materials such as SBT, and high dielectric constant films of materials such as BST.
It is yet another object of the present invention to provide an efficient liquid delivery process for forming Group II metal-containing films such as BST films.
Other objects and advantages of the present invention will be more fully apparent from the ensuing disclosure and appended claims.